Published on Sat Jul 10 2021

End-to-end computational approach to the design of RNA biosensors for miRNA biomarkers of cervical cancer

Baabu, P. R. S., Srinivasan, S., Nagarajan, S., Muthamilselvan, S., Suresh, R. R., Selvi, T., Palaniappan, A.

Cervical cancer accounts for the second largest burden among cancer patients worldwide with an unforgiving 50 percent mortality rate. Poor awareness and access to effective diagnosis have led to this enormous disease burden, calling for point-of-care diagnosis methods.

1
7
31
Abstract

Cervical cancer is a global public health subject as it affects women in the reproductive ages, and accounts for the second largest burden among cancer patients worldwide with an unforgiving 50 percent mortality rate. Poor awareness and access to effective diagnosis have led to this enormous disease burden, calling for point-of-care, minimally invasive diagnosis methods. Here, an end to end quantitative approach for a new kind of diagnosis has been developed, comprising identification of biomarkers, design of the sensor, and simulation of the diagnostic circuit. Using miRNA expression data in the public domain, we identified circulating miRNA biomarkers specific to cervical cancer using multi-tier screening. Synthetic riboregulators called toehold switches specific for the biomarker panel were designed. To predict their dynamic range for use in genetic circuits as biosensors, we built a multivariate linear regression model using a generic grammar of the toehold structure, and thermodynamic features derived from RNA secondary structure and interaction. The model yielded good performance with an adjusted R2 = 0.59. Reaction kinetics modelling was performed to predict the sensitivity of the second-generation toehold switches to the miRNA biomarkers. Simulations showed a linear response between 10nM and 100nM before saturation. Our study demonstrates an end-to-end workflow for the efficient design of genetic ciruits geared towards the effective detection of unique genomic signatures that would be increasingly important in today's world. The approach has the potential to direct experimental efforts and minimise costs. All resources including the machine learning toolkit, reaction kinetics simulation, designed toehold sequences, genetic circuits, data, and sbml files for replicating and utilizing our study are provided open-source (under GNU GPLv3 licence).